I've been using Snowflake Copilot for the past two days and have successfully onboarded my entire Data Science team using Copilot. Snowflake copilot is pretty fast and way faster than OpenAi.

Text-to-SQL is likely to be one of the first challenges successfully tackled by…

AI will primarily benefit Energy Companies.

ClinicalTrial AI: I am developing an AI for clinical trials designed to monitor and analyze recent trends and patterns across various diseases and treatments. Here's how it's structured:

1. I'm sourcing data from ClinicalTrials.
2. DuckDB is being used for database access.
3.…

Netflix uses Java and its cool 😎

youtube.com/watch?v=BaUrpq…

ehrapy: a framework for the exploratory and targeted end-to-end analysis of complex electronic health record (EHR) datasets inspired by the biological omics world

- Exploratory and targeted analysis of Electronic Health Records
- Quality control & preprocessing
- Visualization &…

What remains the greatest challenge in Data Engineering - Moving Data from one system to another.

Despite the availability of numerous technologies for this purpose, the cost of data movement is extremely high and continues to rise annually.

Now OpenAI can support up to 10k files for knowledge retrieval.

So do I really need a vector db while using OpenAI API?

Scaling systems for GPU computing is ten times more challenging than scaling systems for CPU computing.

Using vector databases for large-scale search or Matching operations is challenging !

- Different vector databases can produce different search results even when they contain the same data, because of differences in their architecture and indexing methods. This makes it…

SQL is great but I prefer Python for building data applications and R-lang for visualisation.

AI Strategy for Large Enterprises:

1. Develop a robust Data/AI Engineering team responsible for the entire technology stack.

2. Quickly develop and implement POCs, obtaining direct feedback from end-users.

3. Prioritize cost-effectiveness, focusing on developing temporary…

Unlocking the Data Puzzle: Navigating the Complex World of Data in Life Sciences.

After encountering countless challenges with data in the life sciences, I've decided to blog it.

This is the first post on navigating the complex world of data in the lifescience / pharmaceutical…

OBO  (Open Biological and Biomedical Ontologies) Foundry is a great resource to find Ontology for Life science.

'Biomedical ontology is a set of terms and relations that represent entities in a biomedical field and how these entities relate to each other in a human- and…

Applying Data Structures in Life Science Data Analysis: Over the recent weeks, I used two well-known data structures that have shown immense utility in healthcare and Lifescience contexts.

- Trie Data Structure, particularly Suffix Tries for NLP: This structure is instrumental…

Data Governance is crucial in highly regulated sectors such as pharmaceuticals 💊

🔒 The sharing of data requires strict regulations, as unauthorized access could result in significant leaks or even halt clinical trials.
🤖 Column masking is employed to conceal certain sensitive…

Weight loss drugs aka GLP-1 Drugs are dominating every space including heart disease,

The treatment benefit for heart failure with preserved ejection fraction, accounting for more than half of heart failure cases, has now been replicated. The key finding of much less weight loss…

What is the roadmap of Data Engineering?

That is a trick question, there are just roads but no map!

Data Infrastructure Observability is a challenging task, with few effective tools available for it.

Because data issues vary widely and are specific to different domains, data infrastructure engineers need to dedicate time to working with their product or team to determine which…